The present invention relates to nuclear medicine imaging systems, such as positron emission tomography (PET) scanners; and particularly to circuits for processing signals from radiation detectors in such systems.
Positrons are positively charged electrons which are emitted by radionuclides that have been prepared using a cyclotron or other device. These are employed as radioactive tracers called xe2x80x9cradiopharmaceuticalsxe2x80x9d by incorporating them into substances, such as glucose or carbon dioxide. The radiopharmaceuticals are injected into a patient and become involved in such processes as blood flow, glucose metabolism, fatty acids, and protein synthesis.
Positrons are emitted as the radionuclides decay. The positrons travel a very short distance before they encounter an electron, and when this occurs, they are annihilated and converted into two photons, or gamma rays. This annihilation event is characterized by two features which are pertinent to PET scannersxe2x80x94each gamma ray has an energy of 511 keV and the two gamma rays are directed in nearly opposite directions. An image is created by determining the number of such annihilation events at each location within the field of view.
The PET scanner includes one or more rings of detectors which encircle the patient. Each detector includes a scintillator which converts the energy of each 511 keV photon into a flash of light that is sensed by a photomultiplier tube (PMT). Coincidence detection circuits connect to the detectors and record only those photons which are detected simultaneously by two detectors located on opposite sides of the patient. The number of such simultaneous events indicates the number of positron annihilations that occurred along a line joining the two opposing detectors. Within a few minutes hundreds of million of events are recorded to indicate the number of annihilations along lines joining pairs of detectors in the ring. These numbers are employed to reconstruct an image using well known computed tomography techniques.
Upon stimulation by a gamma ray, the scintillators do not emit light instantaneously, instead the light is emitted with an intensity that decays exponentially with time. The energy of the gamma ray is determined by integrating the single from the light sensor over the duration of the light pulse. The duration of that light pulse limits the rate in which the gamma rays can be detected and processed. With reference to FIG. 1 consider two gamma rays that interact with the scintillator at times T1 and T2. The second gamma ray strikes the scintillator at time T2 which occurs during the light pulse produced by the first gamma ray striking the scintillator at time T1. Thus, when the processing circuit integrates the signal from the photomultiplier the portion of the signal from the gamma ray which occurs after time T2 (indicated by the crosshatched area on the drawing) will be integrated along with the signal produced by the second gamma ray striking the scintillator. Thus, the signals from the two gamma rays will xe2x80x9cpile upxe2x80x9d and not be processed correctly as individual gamma ray invents. Thus, it will appear as though the second gamma ray occurrence has a much greater signal.
Several methods have been developed to produce and correct the effect of pile up on the process signal. One such method is referred to as a variable integration with pulse tail extrapolation. In this method, the integration of the first pulse is stopped when the second pulse occurs. The integration value of the first pulse and the time between the first pulse and the second pulse is used to calculate a correction to the integrated value for the first gamma ray pulse because of the shorter integration time that is used. A correction for the portion of the first pulse is included with the integral of the second pulse. However, the circuitry required for this method is often too complex and expensive for practical application on imaging systems that contain a large number of light sensors and signal processing channels.
Another method of reducing pile up was described by E. Tanaka et al. entitled xe2x80x9cVariable Sampling-Time For Improving Count Rate Performance Of Scintillator Detectorsxe2x80x9d, Nuclear Instruments And Methods158, 1989, pp. 459-466. This method generates a delayed, inverted and attenuated reflection of the original detector signal by means of analog delay line. The attenuation of the reflected pulse is chosen such that:
Ar(t)=xe2x88x92Ao(txe2x88x92td) exp(xe2x88x92td/Tc)
where Ar(t) is the amplitude of the reflected pulse, Ao(txe2x88x92Td) is the amplitude of the original signal, Td is the delay between the original signal and the reflected signal, and Tc is a decay time constant of the scintillator. The original and reflected pulses then are summed to give a clipped pulse as in an output signal as shown in FIG. 2. The output signal is given by the expression:
As(t)=Ao(T)+Ar(T)
since Ao(T) is given by:                                           Ao            ⁡                          (              T              )                                =          0                                      t           less than           0                                                          Ao            ⁡                          (              T              )                                =                                    Ao              ⁡                              (                                  t                  =                  0                                )                                      ⁢                          exp              ⁡                              (                                                      -                    t                                    /                  Tc                                )                                                                          t           greater than           0                      then                                          As            ⁡                          (              t              )                                =          0                                      t           less than           0                                                          As            ⁡                          (              t              )                                =                      Ao            ⁡                          (              t              )                                                            0           less than           t           less than           Td                                                          As            ⁡                          (              t              )                                =          0                                      t           less than           Td                    
FIGS. 3 and 4 show pre-amplifier circuits that use the delay line clipping method to shorten signals from a scintillator and photomultiplier tube (PMT). In FIG. 3, current from the photomultiplier is dropped across resistor RA to generate a voltage signal which is amplified by amplifier A1 that acts as an input buffer for the clipping circuit. The clipping circuit consists of a load resistor RB, a delay line DL and a terminating resistor RC. The value of the load resistor RL is chosen to equal the characteristic impedance of the delay line DL and the value of resistor RC is chosen so that the reflected signal has the correct amplitude. A second amplifier A2 functions as a driver to isolate the impedance of the output cable from the clipping circuit. Resistor RD is chosen to match the characteristic impedance of the output cable. This circuit has a drawback in that it requires two high performance amplifiers.
As an alternative, the circuit in FIG. 4 eliminates one of the amplifiers in the previous circuit by taking advantage of the high impedance output of the photomultiplier tube. The clipping circuit again consists of a load resistor RE, a delay line DL and its terminating resistor RF. The value of load resistor RE is chosen to equal the characteristic impedance of the delay line and the value of resistor RF is chosen so that the reflective signal has the correct amplitude. The current from the photomultiplier tube is dropped across the load resistance RE and the delay line which give an equivalent resistance of one-half RE. The sole amplifier A1 amplifies the voltage signal and acts as a driver to isolate the impedance of the output cable. Although this latter utilizes only a signal amplifier, it has the drawback that the impedance used to convert the photomultiplier signal to a voltage depends upon the impedance of the delay line. If a standard low impedance delay line is employed, the gain required in the amplifier can become very large which may cause performance and stability problems. If the gain of the amplifier is kept low, a high impedance delay line must be used which are relatively large, have limited availability and are relatively expensive.
A circuit for clipping pulses in a signal from a radiation detector such as one that is incorporated in a nuclear medicine imaging system. That circuit has a first node connected to the radiation detector for receiving the signal, a second node, and a third node to which a reference voltage level is applied. In the preferred embodiment of the present invention the third node is coupled to circuit ground. An impedance element is connected between the first and second nodes to provide the desired total load to the PMT. As used herein, an impedance element is an electrical circuit component that has more than a negligible impedance, for example a conductor would not be considered as an impedance element.
A signal delay element, such as an analog delay line for example, has an input connected to the second node and has an output that is coupled to the third node. In the preferred embodiment, another impedance element couples the delay line output to the third node. A load element is connected between the second node and the third node, and has an impedance that matches the impedance of the signal delay element. An amplifier includes a signal input that is connected to the first node and includes a signal output at which is produced the output signal of the circuit for clipping pulses.